As an important part of the manufacturing industry, complex electromechanical equipment plays a crucial role in accelerating the digitalization and intelligent transformation of the manufacturing industry. Among the many digitization technologies, the virtual and real interactivity and full life cycle of digital twin (DT) make it the primary choice for equipment digitization. The key prerequisite and foundation for realizing equipment digital twin (EDT) is to construct a highly credible and accurate twin model, which is difficult to model due to the strong nonlinearity of electromechanical equipment and the lack of clarity in its underlying mechanisms. To address these challenges, this paper proposes a data-driven modeling method based on TCN-Transformer, which enhances the modeling accuracy of complex electromechanical equipment and reinforces the credibility of the twin. Finally, this paper validates the feasibility and effectiveness of the proposed EDT modeling approach using a quadrotor UAV as the experimental subject.